Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[LowVis] graphics contrast is unmentioned (except as exceptions) #96

Closed
joshueoconnor opened this issue May 31, 2015 · 33 comments
Closed

Comments

@joshueoconnor
Copy link
Contributor

Name: Chaals McCathie Nevile
Email: [email protected]
Affiliation: Yandex
Document: W2
Item Number: Guideline 1.4: Make it easier for users to see and hear content...
Part of Item:
Comment Type: technical
Summary of Issue: graphics contrast is unmentioned (except as exceptions)
Comment (Including rationale for any proposed change):
None of the contrast requirements in guideline 1.4 apply to graphics, except where specific types of graphics are specifically exempted.

This fails to account for primarily graphic applications, such as maps, traffic indication, wayfinding tools, and many others.

People have a primarily visual interaction with the Web. Where graphics are presented as a key part of an interface, even given a text-based alternative, many people will rely on the graphics.

It is therefore important to require sufficient accessibility of graphics content for visual users.

Proposed Change:
Add requirements for appropriate visual contrast in graphics used to provide functionality of a site or application.

@joshueoconnor
Copy link
Contributor Author

@joshueoconnor joshueoconnor changed the title graphics contrast is unmentioned (except as exceptions) [LowVis] graphics contrast is unmentioned (except as exceptions) May 31, 2015
@yatil yatil assigned WayneEDick and unassigned WobleDuck Jul 7, 2015
@WayneEDick
Copy link
Contributor

This extension is critical. Most people with low vision have a narrower contrast tolerance. Graphics often favor visual design over separation of figure from ground. Since 1.4 specifically call out "distinguishable" as its purpose, so extending the 1.4 contrast success criteria to is necessary. For active images this can save a [zoom operation/ return to normal] operation.

@WayneEDick
Copy link
Contributor

More: This is extremely subtle. Is an iconic for a item an image representing text or is it just an image? Is the icon at an abbreviation? My intuition to both questions is no, but icons are used as hieroglyphs for words or phrases. I believe that icons that are not text and exceed the scope of 1.4.3 and 1.4.6, but this is an accessibility gap in the guidelines. No change recommended to Understanding WCAG or Techniques. May be a future extension.

@WobleDuck
Copy link

Michael,
I am not quite up to speed on the github stuff. Do I just post them and you
moderate, or is there a way to just pop them to the list when I'm done?

Wayne

On Tue, Jul 7, 2015 at 8:31 AM, michael-n-cooper [email protected]
wrote:

Assigned #96 #96 to @WobleDuck
https://github.com/WobleDuck.


Reply to this email directly or view it on GitHub
#96 (comment).

@joshueoconnor
Copy link
Contributor Author

@WayneEDick Please post any suggestions, drafts, comments or questions here in the relevant thread. You can add things to the wiki also, and add the URI here, if that helps. Thanks

@WayneEDick
Copy link
Contributor

I have one additional thought on this. Iconic Fonts should be a different story. Use of Iconic fonts that do not include enough internal contrast or do not contrast enough from the background should be a failure. These are being used as letters and should be treated like alphabetic objects. 1.4.3 and 1.4.6 apply.

@WobleDuck
Copy link

I done with this without more comment.

On Sat, Sep 12, 2015 at 2:30 PM, WayneEDick [email protected]
wrote:

I have one additional thought on this. Iconic Fonts should be a different
story. Use of Iconic fonts that do not include enough internal contrast or
do not contrast enough from the background should be a failure. These are
being used as letters and should be treated like alphabetic objects. 1.4.3
and 1.4.6 apply.


Reply to this email directly or view it on GitHub
#96 (comment).

@WayneEDick
Copy link
Contributor

There is one issue that should be addressed. If the contrast ratio is too low for anyone with low vision to see then it should simply be accompanied by a text alternative. Below 4.5:1 is pretty low, so this would qualify in most cases. Currently the only way to read with text alternatives is with screen magnifiers and screen readers, both of which do not serve the population adequately, but the real solution is to improve the technology for delivering alt-text to people with low vision.

@kwahlbin
Copy link
Contributor

kwahlbin commented Jan 4, 2016

Images that convey information and are not accompanying by text should have a 3:1 color contrast ratio. The challenge here is if the images have multiple colors. The image needs to be differentiated from other icons on the page and one of the colors in the icon need to meet the 3:1 color contrast ratio.

@srinivasuchakravarthula
Copy link

+1 to Kathy. Unless they are photographs (natural), it's advisable to use right set of colors while designing images if there is an intention to embed text on it.

@MikeElledge
Copy link

Wayne, do you agree that icons without text should have a 3:1 contrast ratio? I've been using 4.5:1 for our icons, which limits the color pallet, but is quite readable.

@lauracarlson
Copy link
Contributor

As stated in the November 20, 2015 Accessibility Requirements for Low Vision Users Editor's Draft:

2.1.3 Contrast Adjacent

Sometimes adjacent colors need to be distinguished, for example, colors next to each other in a pie chart. Such colors should have sufficient contrast and be distinguishable by people who are color blind and have reduced contrast sensitivity.

User Requirement:

Users can distinguish between adjacent colors when needed for meaning.

@chaals does this Adjacent Color User Requirement help to cover what you intended in your initial comment? What else did you have in mind? Ideas for improvement?

@awkawk
Copy link
Member

awkawk commented Jan 5, 2016

1.4.3: The visual presentation of text and images of text has a contrast ratio of at least 4.5:1, except for the following: (Level AA)

What is not covered currently is graphic information that is not text. I think that we need to defer the gap to the LVTF to address.

@michael-n-cooper
Copy link
Member

I see separate threads here:

  1. Extending contrast requirements currently applicable only to text, to graphics;
  2. Changing contrast threshold required for conformance;
  3. Classifying graphics to specify what contrast requirements might be applicable, e.g., text-in-images vs. photographs that don't primarily convey text meaning, perhaps others.

All of those are in scope. But they should be handled separately. Otherwise you end up with requirements that are too specific to certain use cases considered by the group, not generalized, and might also have too much divergence from the core guidelines.

@DavidMacDonald
Copy link
Contributor

If text that is part of the page content is in a graphic, 1.4.3 applies to that text

@MikeElledge
Copy link

I agree that this should be forwarded to the LVTF so they can recommend appropriate criteria.

@WobleDuck
Copy link

Wobble Duck is me. I posted this on the use cases of LVTF

On Tue, Jul 7, 2015 at 8:31 AM, michael-n-cooper [email protected]
wrote:

Assigned #96 #96 to @WobleDuck
https://github.com/WobleDuck.


Reply to this email directly or view it on GitHub
#96 (comment).

@chaals
Copy link
Contributor

chaals commented Jan 21, 2016

I agree with Michael that there are several threads here.

The point is not about adjacent colours - anti-aliasing and so on can confuse this endlessly, but that the important features of a graphic should be distinguishable without relying on high-contrast separation.

Thus icon fonts need to use something better than two shades of grey - but also something clearer than a single row of pixels...

@WayneEDick
Copy link
Contributor

We will address this in the LVTF.

It is clear that icons are being used like hieroglyphs and so the
requirement of visibility is more urgent. It is time to collect a
vocabulary of icons and insist on visible (1.4.3) and consistent (3.2.3,
3.2.4) usage. I am sure there will be a recommendation regarding these
items because the usage is more linguistic than graphic.

Wayne

On Thu, Jan 21, 2016 at 2:59 AM, chaals [email protected] wrote:

I agree with Michael that there are several threads here.

The point is not about adjacent colours - anti-aliasing and so on can
confuse this endlessly, but that the important features of a graphic should
be distinguishable without relying on high-contrast separation.

Thus icon fonts need to use something better than two shades of grey - but
also something clearer than a single row of pixels...


Reply to this email directly or view it on GitHub
#96 (comment).

@joshueoconnor
Copy link
Contributor Author

@lauracarlson has kindly agreed to draft a technique that outlines requirements for appropriate visual contrast in graphics used to provide functionality (in a website or application).

@lauracarlson
Copy link
Contributor

I drafted a Using sufficient contrast for images that convey information technique.

Ideas for improvement are welcome and appreciated. cc/@chaals

Thanks!

@awkawk
Copy link
Member

awkawk commented Apr 12, 2016

Preparing to mark as deferred due to inactivity and group resource constraints (https://www.w3.org/WAI/GL/wiki/Handling_Issues).

@lauracarlson
Copy link
Contributor

@chaals any thoughts on the draft Using sufficient contrast for images that convey information technique?

@lauracarlson
Copy link
Contributor

Hi @joshueoconnor and @awkawk,

I worked on a technique in the Wiki last February. It is: Using sufficient contrast for images that convey information.

Does that qualify as activity? Thoughts?

Thanks,
Laura

@joshueoconnor
Copy link
Contributor Author

@lauracarlson it sure does :-) I'm marking it as 'ready for survey' (again as it already was but looks llike it wasnt surveyed.

@joshueoconnor
Copy link
Contributor Author

@lauracarlson - we did survey this right? And IIRC, there was some more work/edits to be done?

@lauracarlson
Copy link
Contributor

@joshueoconnor Yes. I also need to consult with the LVTF.

@lauracarlson
Copy link
Contributor

@joshueoconnor
Copy link
Contributor Author

@lauracarlson Thanks for the new proposed SC. Is this ready for survey?

@lauracarlson
Copy link
Contributor

lauracarlson commented Aug 16, 2016

Hi @joshueoconnor,

I've asked the LVTF. I'm wondering if it should be put into the SC template along with the rest of the new LVTF SCs for 2.1.

@lauracarlson
Copy link
Contributor

lauracarlson commented Aug 18, 2016

@lauracarlson
Copy link
Contributor

lauracarlson commented Sep 6, 2016

@joshueoconnor
Copy link
Contributor Author

Taken up by LVTF - closing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests